AI for Climate
Machine learning applied to weather and climate forecasting, Earth observation, energy grid optimization, and materials discovery for batteries and electrolyzers — the computational frontier of climate response.
Geospatial foundation models are a new generation of artificial intelligence that learn to understand vast amounts of satellite imagery and other location based data. These models can then apply this understanding to a wide range of new tasks, from predicting crop yields to mapping flood zones, without needing extensive task specific training. The 126 papers published between 2024 and 2026 show this field is rapidly maturing, moving beyond initial research to practical application and solving the significant challenge of extracting actionable insights from Earth's complex and ever changing landscape.
Pioneer teams are investing seriously. Methods are clarifying and early results are compelling. This is when category leaders typically emerge.
Within the next 12 to 18 months, expect to see early adopters in agriculture and disaster response begin deploying these models for more accurate, real time monitoring. By 2027 to 2028, as the technology becomes more accessible and benchmarks like GEO-Bench 2 mature, broader adoption across urban planning and environmental consulting will occur, driven by imitation as businesses see competitors gain an advantage. The peak of activity is projected in about three years, but challenges remain in ensuring these models are trustworthy and can handle real world data degradations, as highlighted in papers like RSPDBench.
Get next month's AI for Climate movement in your inbox, free.
Full geographic breakdown with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Research Signals · 6 clusters detected
Climate Downscaling
? What if we could predict precisely how a specific neighborhood will be impacted by extreme weather a decade from now, allowing businesses and governments to make incredibly targeted and proactive investments to protect communities and infrastructure
This research cluster explores a new generation of artificial intelligence, called "foundation models," that can understand and predict climate at much finer scales than previously possible. By analyzing over 100 recent research papers, we see a significant push to create AI systems that learn from vast amounts of Earth observation data and climate model simulations, enabling more precise local climate forecasts. This is critical for tackling the increasing impacts of climate change, from extreme weather events to long-term shifts in regional climates.
Key Papers
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundTrajectories for the other 5 sub-clusters with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Deep Analysis
In the next 1-2 years, expect early adopters in sectors like agriculture and insurance to begin using these AI tools for more accurate seasonal weather predictions and risk assessments. Over the next 3-5 years, as these models mature, we will see broader adoption across industries for climate-resilient infrastructure planning and resource management. With a long growth runway ahead, reaching peak development in about 21 years, the primary barrier remains the computational cost and the need for even more diverse and high-resolution training data.
PEST Implications
Governments can use this research to better predict and prepare for the local impacts of climate change, informing policy decisions on infrastructure and disaster relief.
Businesses can leverage more accurate local climate predictions to manage risks, optimize resource allocation, and develop new climate-resilient products and services.
Communities can benefit from improved understanding of local climate risks, enabling better planning for public health, agriculture, and water management.
This research is rapidly advancing the tools and methods for understanding climate at a local level, leading to more powerful and accurate climate simulations.
Deeper signals with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Geospatial Foundation Models
? What if you could instantly see the hidden risks and opportunities in any corner of the world, from where to build the next profitable factory to how to save communities from natural disasters, all powered by AI that understands our planet like never before
Geospatial foundation models are a new generation of artificial intelligence that learn to understand vast amounts of satellite imagery and other location based data. These models can then apply this understanding to a wide range of new tasks, from predicting crop yields to mapping flood zones, without needing extensive task specific training. The 126 papers published between 2024 and 2026 show this field is rapidly maturing, moving beyond initial research to practical application and solving the significant challenge of extracting actionable insights from Earth's complex and ever changing landscape.
Key Papers
+ 5 more papers with Pro
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundCrossing into mainstream, this is the fastest-moving window before the field commoditises. Research intensity peaks around 2029.
Hover the chart to explore adoption over time
Phase Position
Deep Analysis
Within the next 12 to 18 months, expect to see early adopters in agriculture and disaster response begin deploying these models for more accurate, real time monitoring. By 2027 to 2028, as the technology becomes more accessible and benchmarks like GEO-Bench 2 mature, broader adoption across urban planning and environmental consulting will occur, driven by imitation as businesses see competitors gain an advantage. The peak of activity is projected in about three years, but challenges remain in ensuring these models are trustworthy and can handle real world data degradations, as highlighted in papers like RSPDBench.
PEST Implications
Governments can use these advanced mapping tools to better monitor environmental changes and plan for natural disaster response.
Businesses can gain a competitive edge by using these models to predict market trends or optimize resource allocation based on location.
Improved understanding of local environments through these technologies can help communities adapt to climate impacts and enhance urban planning.
This research is rapidly advancing how computers understand and process vast amounts of location-based information, leading to smarter applications.
Deeper Signals
Reads as Early Adopters based on how these papers are written, not the Early Majority stage the paper-volume model above shows. Papers frequently use terms like "propose," "explore," "potential," and "achieving state-of-the-art," indicating a focus on research and development rather than widespread production deployment.
Weather Forecasting Models
? What if businesses could predict extreme weather events with pinpoint accuracy days in advance, averting billions in losses and safeguarding communities
New artificial intelligence (AI) models are rapidly transforming weather forecasting, moving beyond traditional physics-based simulations to achieve comparable or even superior accuracy with significantly less computational power. The sheer volume of 114 research papers published between 2023 and 2026 signals a robust wave of innovation, focused on developing more precise, faster, and increasingly probabilistic weather predictions. This research unlocks the capability for highly detailed, near real-time forecasts that can anticipate extreme events and complex weather patterns with unprecedented reliability.
Key Papers
+ 5 more papers with Pro
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundTrajectories for the other 5 sub-clusters with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Deep Analysis
In the next one to two years, early adopters in sectors like aviation and energy will integrate these AI forecasts for operational decision-making, while research continues to refine uncertainty quantification. Over the next three to five years, as adoption grows, these advanced models will become standard tools for agricultural planning, disaster management, and even urban infrastructure planning, driven by their increasing accuracy and accessibility. With a long growth runway ahead, peaking in approximately 22 years, this field will continue to evolve, pushing the boundaries of forecasting into subseasonal and seasonal predictions.
PEST Implications
Governments can use these advanced weather predictions to better prepare for and respond to extreme weather events, protecting citizens and infrastructure.
Businesses relying on weather, like agriculture or energy, can make more profitable decisions by anticipating future conditions more accurately.
Improved weather forecasts help individuals plan daily activities and stay safer during severe weather, enhancing community well-being.
New AI and computing methods are revolutionizing how we predict weather, leading to faster and more precise forecasts than ever before.
Deeper signals with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Remote Sensing Segmentation
? What if we could instantly know the precise location and condition of every valuable resource or environmental change on the planet just by looking at a satellite image?
This research uses advanced AI to automatically identify and map features in satellite images, enabling better land management and resource monitoring.
Key Papers
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundTrajectories for the other 5 sub-clusters with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
PEST Implications
Governments can use this technology to monitor borders and track environmental changes for better policy making.
Businesses can improve resource management and optimize logistics by accurately mapping land use and identifying valuable assets from above.
Improved environmental monitoring can help communities prepare for natural disasters and understand the impact of climate change on their surroundings.
This research is rapidly advancing the ability of computers to automatically understand and interpret images from satellites and drones, leading to new applications.
Deeper signals with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Energy System Optimization
? What if we could predict and instantly fix every glitch in our power grid before you even notice the lights flicker, ensuring uninterrupted, clean energy for your business and home?
This research optimizes energy systems using AI to improve reliability, efficiency, and sustainability, crucial for stable and green power delivery.
Key Papers
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundTrajectories for the other 5 sub-clusters with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
PEST Implications
Governments may need to update regulations as new energy technologies emerge, potentially impacting national energy security and international agreements.
Businesses can find new opportunities in developing and implementing advanced energy solutions, leading to cost savings and new market competitiveness.
Communities could experience changes in energy access and reliability, influencing local development and environmental quality.
Expect significant advancements in how we generate, distribute, and consume energy, making systems smarter and more efficient.
Deeper signals with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Earth Observation Learning
? What if we could predict crop yields, natural disaster impacts, or resource availability with uncanny accuracy simply by analyzing everyday satellite images, revolutionizing how we plan and invest in our planet's future
This research uses AI to understand satellite images for better environmental monitoring and resource management, leading to smarter business decisions.
Key Papers
+ 5 more papers with Pro
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundPhase Position
Adoption Curve
Floor, not a measurement · monthly rate flat, no peak yet — no ceiling estimable, so this is a lower boundTrajectories for the other 5 sub-clusters with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
PEST Implications
Governments can use this technology to better monitor borders and natural disasters, improving national security and disaster response.
Businesses can gain insights into crop yields and land use, leading to more efficient agricultural practices and resource management.
Improved mapping of vulnerable areas can help communities prepare for and mitigate the impact of climate change and natural hazards.
This research is rapidly advancing the ability of computers to understand complex satellite imagery, opening doors for new applications in environmental monitoring and urban planning.
Deeper signals with Pro
Adoption trajectories · Convergence scores · Geographic splits · Archive
Unlock the data · $39 / monthCancel anytime · Annual plan $390/yr
Research Momentum
Based on representative paper sample per cluster · not a complete count
Share of papers per adoption stage, weighted by cluster size.
Related Trends · AI/ML
Research insights powered by Google Gemini
Manage subscriptionContactMCP for ClaudeTermsPrivacy
Built by Donald Butts